Repetition and associative context effects in speech production

An experiment is reported that addresses the impact of lexical constraint on the production characteristics (mean duration and peak amplitude) of a given target word within a sentence production task. Subjects produced aloud sentences from memory that contained a target word (e.g., cat) that was either a repetition of an earlier word in the sentence (e.g., cat), associatively related to an earlier word in the sentence (e.g., dog), or unrelated to an earlier word in the sentence (e.g., son). The position within the sentence and phonetic environment of the target were equated across conditions. The results indicated that the mean durations for the target word were shorter in the repetition condition, compared to the associatively related condition, which in turn produced shorter production durations compared to the unrelated condition. In addition, the peak amplitude measurements indicated that the repeated condition produced relatively lower peak amplitudes for the target word, compared to the remaining two conditions which did not differ.